Oxygen's Role in Earth's Atmosphere: Production, Isotopes, and History, Lecture notes of Natural History

Download Oxygen's Role in Earth's Atmosphere: Production, Isotopes, and History and more Lecture notes Natural History in PDF only on Docsity! The Natural History of Oxygen MALCOLM DOLE From the Department of Chemistry and the Materials Research Center, Northwestern University, Evanston ABSTRACT The nuclear reactions occurring in the cores of stars which are believed to produce the element oxygen are first described. Evidence for the absence of free oxygen in the early atmosphere of the earth is reviewed. Mech- anisms of creation of atmospheric oxygen by photochemical processes are then discussed in detail. Uncertainty regarding the rate of diffusion of water vapor through the cold trap at 70 km altitude in calculating the rate of the photo- chemical production of oxygen is avoided by using data for the concentration of hydrogen atoms at 90 km obtained from the Meinel OH absorption bands. It is estimated that the present atmospheric oxygen content could have been produced five to ten times during the earth's history. It is shown that the isotopic composition of atmospheric oxygen is not that of photosynthetic oxygen. The fractionation of oxygen isotopes by organic respiration and oxidation occurs in a direction to enhance the O is content of the atmosphere and compensates for the 018 dilution resulting from photosynthetic oxygen. Thus, an oxygen isotope cycle exists in nature. I. I N T R O D U C T I O N Although it is estimated that oxygen is only the third most abundant element cosmically (1) coming behind hydrogen and helium, in that order, it is the most abundant element on the earth's crust, which is important for the living race as oxygen is essential for life. As we shall see later, oxygen was very probably not present in the early atmosphere of the earth, so it is interesting to consider how oxygen arose in the atmosphere, whether its abundance is now changing, what its isotopic composition is, and how the latter varies between samples of oxygen from different sources. Furthermore, there is probably a close connection between the origin of life on this planet and the growth in the abundance of atmos- pheric oxygen as recently emphasized by Berkner and Marshall (2). Before con- sidering the interesting factors involved in the development of atmospheric oxygen, however, we shall take up first the history of the formation of the element oxygen in the cosmos. It should be emphasized that many of the con- clusions described below must be considered not in the same light in which we view well grounded scientific laws and principles, bu t rather as the best The Journal of General Physiology STRUCTURE AND FUNCTION OF OXYGEN intelligent guesses and deductions that we can now make on the basis of presently available facts and theory. II. C O S M I C F O R M A T I O N O F O X Y G E N It is believed that the energy radiated by a star throughout most of its life comes from the so called hydrogen-burning nuclear reaction in which hy- drogen is converted to helium (3). If this is true, then at zero time the universe must have consisted largely and probably solely of hydrogen, because there are few nuclear reactions which spontaneously produce hydrogen. In other words, hydrogen is continually being converted to helium and heavier ele- ments and because it is the most abundant element in the universe today it must have been present in overwhelming amount and very possibly completely pure at zero time. The hydrogen-burning reaction in which helium is produced is the most efficient energy producer, but the actual mechanism probably depends on the temperature of the star. At relatively low temperatures and in the early stages of a star's life before much He 4 has been produced, the hydrogen- burning chain reaction consists of the following steps: 1H 1 + 1H 1 - - o l H ~ + / 3 + + v where/3 + represents the positron and v the neutrino, (1) 1H 2 + 1H 1 --~ ~He 3 + "y ( 2 ) and ~He ~ + 2He 3 ~ H e 4 + 2 1 H 1 ( 3 ) where 3' represents a gamma ray photon. The total energy yield is 26.2 Mev (equivalent to 6 X l0 s kcal per mole of 2He 4) for the net process which is 41H J --o 2He 4 + 2~3+ + 2v + 23/ (4 ) (There is a 2 per cent energy loss due to the production of the neutrino.) There are other mechanisms for He formation but the above is believed to be the most important. As the hydrogen is consumed in the core of the star to form helium, no further nuclear transformations can take place until both the temperature and density have greatly increased. At temperatures of 108°K and densities of 105 gm cm -3 sC 12 can be produced by the nuclear reaction 32He 4 ~ ~C t2* --o 6C '2 + T (5) with an energy release of 7.3 Mev per a tom of ~C '2. This reaction probably MALCOLM DOLE Natural History of Oxygen 9 in which 1 mole of the product oxygen is chemically equivalent to 1 mole of the reactant carbon dioxide. If the carbohydrates formed in the reaction eventually are reduced to coal or fossil fuels, then the sum of the carbon in the latter should be chemically equivalent to the free oxygen of the atmos- phere. Actually a considerable excess of carbon is now recognized, hence there is no necessity for assuming the presence of any free oxygen in the initial atmosphere of the earth. In fact, Rubey (10) has estimated that the entire carbon-oxygen ratio of the earth's crust, atmosphere, hydro-, and biospheres, can be accounted for by an approximately 3 to 2 ratio of CO2 and CO "which is within the range of occluded CO2 and CO proportions actually found in igneous rocks." The reason that we made the statement above that all of the oxygen of the present atmosphere was produced by the photosynthetic reaction is that only approximately 2000 years are required to produce all of the atmospheric oxygen (2). This is a very short time on the geologic time scale. Two other arguments against the existence of free atmospheric oxygen in the primeval atmosphere are the following: It is generally assumed that life on this planet could not have arisen in an oxidizing atmosphere. For example, Miller (11) was able to synthesize amino acids by passing an electric spark through a mixture of CH4, NH3, H2, and H~O, but no organic compounds were formed when the gas mixture was only CO2 and H20. Abelson (12) found that CH4 -b N2 + H20 gave amino acids in an electric discharge, but not when COs -I- N~ -b H20 was the gas mixture. In the presence of oxygen and light, especially ultraviolet light, the lifetime of amino acids would be very short. Thus, these authors believe that for life to have started, free oxygen must have been absent from the early atmosphere. The second reason for the absence of oxygen in the early atmosphere stems from the deduction of Brown (13) that because most of the neon and a large fraction of the other inert gases escaped from the earth during its formation from its protoplanet, no free gas could have existed in the atmosphere. Only those gases were retained that could be retained chemically, H20, CO2, 02, and N2. For oxygen to exist as free gaseous oxygen it had to be liberated from chemical combination. Possible mechanisms for such liberation via photo- chemical means are discussed in the next section. IV. T H E P H O T O C H E M I C A L T H E O R Y O F T H E F O R M A T I O N O F A T M O S P H E R I C O X Y G E N A. Photochemical Reactions Although the greatest energy of the sunlight striking the earth per 50 A range of wavelength is in the visible (2) at about 4500 A wavelength, nevertheless, there is considerable energy in the ultraviolet region of the spectrum. As the wavelength is decreased to 2000 A or lower, water and then carbon dioxide IO S T R U C T U R E A N D F U N C T I O N O F O X Y G E N begin to absorb the ultraviolet light. At 2000 A the photons have enough energy on absorption to dissociate water vapor according to the reaction H20 + hv ~ H + OH +5.12 (11) (The energies of reactions will be given in units of electron volts per event; a plus sign signifies energy absorbed. Energies of many reactions have been collected by the author (14). hv represents the photon.) The hydroxyl radical, OH, produced in reaction (11) can also absorb at even longer wavelengths than the water molecule itself and undergo further decomposition according to the reaction OH + h v - * O + H +4.34 (12) Atomic oxygen formed in this reaction can unite with another atom of oxygen in the presence of a third body such as a nitrogen molecule to form molecular oxygen O + O N~O~ --5.08 (13) Thus, it is not difficult to understand how molecular oxygen could have arisen in the earliest times. There are serious problems, however. In the presence of ultraviolet light, oxygen also absorbs (see Fig. 1) to form atomic oxygen which then reacts in the presence of a third body to produce ozone according to the reactions O3 + h u g O + O +5.08 (14) 02 + O N2 03 --1.03 (15) The ozone absorbs ultraviolet light up to considerably higher wavelengths than does oxygen or water. This energy absorption produces electronically excited ozone molecules which can become deactivated by collisions, or which decompose to atomic and molecular oxygen, or which can react with H or H~, viz. Oa + h g ~ O * (16) O* N,, 03 (17) o* --, o,. + o (18) O* + H~ ---* OH + HO2 (19) o r --~ OH + H + 02 (20) MALCOLM DOLE Natural History of Oxygen z I Further reaction of these radicals would lead to water formation. The yield of the reaction is one-half molecule of hydrogen reacted per quan tum of light absorbed by the ozone (15). For each molecule of water formed one atom of free oxygen is lost and reactions (11) and (12) essentially reversed. Obviously, for oxygen to build up in the atmosphere it is necessary for hydrogen to escape into outer space. The author some time ago (14) con- sidered many possible reactions that might trap atomic hydrogen and pre- ,0 2 =E o I 0 1,.-, z __.1 t l . o o i O - I z o I - o m ~ 10-2 1003 I000 4500 10 2 H20 o~ A,o [ A T o 10-5 I I I I I I I I ! I I " "~ , I 1200 1400 1600 1800 2 0 0 0 2 200 WAVELENGTH (A} Floum~ l. Ultraviolet absorption coefficients at different wavelengths for some atmo- spheric gases from Berkner and Marshall (2). Figure reprinted by permission of the Faraday Society from Discussions o/the Faraday Society, 1964, 37, 122. vent it from escaping, but he could find none that looked very plausible. Even at heights in the atmosphere where both ozone (16) and atomic hydrogen can be produced, about 90 km, the following reactions are believed to occur (17) Oa + H --~ O2 + OH --3.30 OH + O ~ O 2 + H --0.73 (21) (22) The net result is simply Os + O --* 2 O2 leaving the H atom concentration unchanged. Evidence for these reactions is found in the Meinel O H bands at I4 STRUCTURE AND FUNCTION OF OXYGEN mate of the water vapor pressure considerably lower than that of Harteck or Kuiper, he deduced a rate of H a tom escape about 500 times less than the rates given above. However, any turbulence or mixing in the stratosphere would raise Urey's estimate. But the whole problem of the partial pressure of water vapor can be avoided by considering only the rate of H a tom escape as explained below. Although some of the H atoms of reaction (2 I) may have come from the photodissociation of CH4, the low abundance of the latter and the probabili ty of its reaction with ozone or nitrogen oxides at lower levels indicate that this source of H atoms can be neglected. But for the 02 photochemical process to leave a residue of free atmospheric oxygen, the H atoms must diffuse from the lower cold levels to higher hot levels where their kinetic energy will be sufficient for escape. Using equation T A B L E I I I R E S U L T S O F ESCAPE AND D I F F U S I O N C A L C U L A T I O N S (Assuming H a tom concen t ra t ion at 90 km to be 1.5 X I0 ~ a t o m s / c m 8) H a t o m escape Conditions No. of H atoms Yrs. to reproduce O~ in Height Temperature escaping/cmS sec. present atmosphere 90 km 200°K 0.15 4X1018 125 500 5 .80X 108 0 .95X 109 Wate r diffusion calculat ions 20 260 (Har teck and 0 .83X 108 Jensen , 1948) 40 270 (Kuiper , 1952) 0.62X109 H a tom diffusion 90 220 (average) 1.5XlO 9 0 .42X 109 Es t imated age of ea r th 5XlO 9 (26) to make this calculation from the 90 km level assuming the H atom concentration to be 1.5 × 10Vcm ~ as before, it turns out that 0.42 × 109 years would be required for enough H atoms to diffuse so that the present atmospheric O2 content could be generated. The results of the diffusion and escape calculations are collected in Table III . If the H atom concentration is 1.5 × 109 a toms/cm 3 at 90 km, the con- centration would probably be somewhat less than this at 125 km; therefore, the calculation of the rate of the H atom escape at 125 km may be too fast by the ratio of these concentrations. Nevertheless, it would appear that these new estimates based on recent determinations of the H atom abundance at 90 km confirm the earlier deductions of Harteck and Jensen and of Kuiper, that the photochemical dissociation of water vapor to oxygen and hydrogen during the lifetime of the earth has been of considerable geological signifi- cance. MALCOLM DOLE Natural History of Oxygen 15 V. O X Y G E N A N D T H E O R I G I N O F L I F E A. Obstades to Life in the Earliest Times Intelligent life as we know it requires an extremely specialized environment whose temperature and composition have to be carefully regulated within rather narrow limits for life to survive, as was pointed out in a masterly fashion many years ago by L. J . Henderson (22). In the earliest times be- cause of less intense radiation from the sun, the earth may have been somewhat colder than it is today (23), in which case the oceans may have either not yet been formed or may have been too cold, i.e. frozen, so that photosynthesis could not have occurred. The initiation of life may have had to await, there- fore, a certain warming up of the earth. Another obstacle to the creation of life in the earliest times must have been the intense ultraviolet radiation striking the earth's surface. In Fig. 1 the absorption coefficients in the ultraviolet are plotted for a number of gases and it can be seen that only ozone has a significant absorption coefficient for wavelengths greater than about 2000 A. In fact, the absorption coefficient of ozone goes through a maximum (24) at 2537 A equal to about 135 cm -1. This means that in the early atmosphere before the high altitude accumulation of ozone, the ultraviolet intensity at the earth's surface must have been great enough to be lethal to many organisms. Especially the combination of oxygen, even ff the latter is present only at low concentrations, with ultraviolet light would be particularly deleterious (12). Berkner and Marshall (2) have re- cently reemphasized the need for a protective ozone layer to shield the earth from the lethal ultraviolet radiation. More importantly, they have shown a possible connection between the various stages in the evolution of life and the levels of oxygen concentration in the atmosphere. It should be noted, how- ever, that Urey (21) has suggested that there may have been organic com- ponents in the early atmosphere of sufficient abundance to have provided a primeval ultraviolet screen. B. The Early Chemistry of Atmospheric Oxygen As soon as the water vapor content of the atmosphere rose to significant levels, i.e. as soon as the earth warmed up if it had been cold, the photochemical process for the production of oxygen must have started. At altitudes of 10 km and above the processes going on would be little affected if at all by the con- dition of the earth's surface. As soon as oxygen was produced it would begin diffusing downwards. As soon as it reached the top of the troposphere, about 10 kin, it would rapidly mix with all the gases of the troposphere because of winds and turbulence in the latter. Because the formation of free atmospheric oxygen depends on the escape of hydrogen from the earth and because the latter may have occurred slowly, x6 STRUCTURE AND FUNCTION OF OXYGEN probably the oxygen that diffused to the earth in the earliest time was en- tirely consumed by oxidation of components of the earth's crust. However, oxygen would build up in the upper atmosphere until its rate of production equalled its rate of diffusion to the earth's surface. If we assume that the rate of production is correctly given by the value calculated from the H atom diffusion, namely 3.7 × 108 molecules of O~/sq. cm sec. (Table I I I ) , then by the use of equation (26) we can calculate the value of co necessary to make J equal to 3.7 × 108. In this calculation we assumed that the diffusion oc- curred over the height from 40 to 10 km, that the average temperature was 270°K, and that oD was equal to 7.5 × 10 .6 moles /cm see. ; co is the ratio of molecules of oxygen to that of nitrogen. The calculation yielded about 3 × 10 -4 for co or about 1.2 × 10 -3 of the present ratio. This abundance of oxygen is very small, and not enough to screen effectively the earth from the ultra- violet rays of the sun. We conclude that for many millions of years there was not enough oxygen in the earth's atmosphere to be of significance in protecting the earth from the lethal ultraviolet sunlight. C. Photosynthetic Production of Oxygen Following Berkner and Marshall (2) it seems evident that the simplest life forms must have arisen in the oceans, perhaps in relatively stagnant bays (example, Phosphorescence Bay off the coast of La Parguera, Puerto Rico), at depths great enough so that the water protected them from the lethal ultraviolet rays, up to 30 feet (2). As soon as algae developed, photosynthesis could begin, thereby greatly accelerating oxygen production. Berkner and Marshall (2) associate the rise of oxygen in the atmosphere to 1 per cent of its present level with the beginning of the Cambrian era, about 550,000,000 years ago, when there was an evolutionary explosion. VI. T H E I S O T O P I C C O M P O S I T I O N O F A T M O S P H E R I C A N D P H O T O S Y N T H E T I C O X Y G E N A. The Isotopic Composition of Various Oxygen Sources Compared The present oxygen of the atmosphere must have been produced by the photo- synthesis reaction, because it takes only about 2000 years to regenerate photo- synthetically all of the free oxygen of the atmosphere. It would therefore be expected that its isotopic composition would be that of photosynthetic oxygen. Very surprisingly it is not. During the first third of this century, one would not have even thought to question oxygen's isotopic composition. In the first place the isotopes of oxygen were not discovered (4) until 1929, and in the second place it was one of the principles of chemistry that no isotope separation could occur in MALCOLM DOLE Natural History o] Oxygen 19 plankton, bac te r ia and o ther sea life which consume oxygen must prefer- ent ia l ly metabol ize 016 at a h igher ra te t han 018 to p roduce this m a r k e d isotope f rac t iona t ion . " Fu r the r work by Dole, Lane , R u d d , and Zaukelies 0.2130 30 o.~mo I~ % Oa -") ' ,o jo < % 0 le / 0.2040~ 0 DEPTH IN METERS I 1 0 0 0 2000 3000 Fmul~ 2. Atom per cent of O is in dissolved ocean air at different depths in the Pacific Ocean at 32 ° 10' N and 120 ° 19' W. Also plotted is the volume percentage of oxygen in the air. (33) conf i rmed these results and led to a m a t h e m a t i c a l descr ip t ion of the d a t a in terms of the following equa t ion 1 -- a l n C 2 = lnY~"2 (30) ot c I y w , l where c2 and ca are the concent ra t ions of oxygen in the ocean wate r a t two di f ferent depths, y~.~ and Y~,,1 are the O I s percentages in the dissolved oxygen at the two depths, and a is the isotope f rac t ionat ion factor def ined by the ra t io y,~/y~ where yB is the average 018 pe rcen tage in the oxygen consumed by all the l iving organisms or ox ida t ion react ions going on in the ocean (as- sumed to be i n d e p e n d e n t of d e p t h and def ined at the same dep th as y,,). A plot of In c versus In y ~ yielded rough ly a s t ra ight line f rom whose slope a was 2 0 S T R U C T U R E A N D F U N C T I O N O F O X Y G E N calculated to be 1.009. There was considerable scatter to the data at low oxy- gen percentages due in part to the difficulty of making accurate O t s per- centage measurements in samples of air containing small concentrations of oxygen. Vinogradov et al. (37) explain the above mentioned scatter in terms of variations in temperature of the water at the four different geographical stations at which the dissolved ocean air was collected. Isotope fractionation factors usually change with the temperature. However, the average value of a found by Vinogradov et al. (37) in their extensive work was 1.010 which is within the limits of error equal to that found by Dole and coworkers, namely 1.009. D. Theories of the Enhanced 018 Content of Atmospheric Oxygen We have seen above that the oxygen produced in the photosynthesis reaction does not have the same isotopic composition as the oxygen already in the air, hence there must be some mechanism which tends to enhance the O 18 abun- dance in air. Calculations show that the atmospheric O 18/O16 ratio cannot be the result of the isotopic exchange equilibrium reaction because the equi- librium constant of the latter is much too close to unity. Inasmuch as the oxygen of the air and the waters of the oceans covering most of the earth's surface are in continual contact, there cannot be any isotope exchange going on between the molecular species H~O and 02, otherwise the equilibrium would be established. In fact, there is no known mechanism by which H20 and O5 can undergo oxygen isotope exchange in aqueous systems at room temperature; high temperatures and catalysts like plat inum are required. Without taking space here to describe other possibilities which we have considered and rejected, let us turn at once to two reasonable possibilities. The fractionation of oxygen isotopes during the consumption of oxygen in the ocean suggested that this mechanism might explain the enhanced O is content of the atmosphere. Accordingly, Lane and Dole (39) investigated tractionation during respiration of a number of organisms, including Homo sapiens. For the small organisms the apparatus used was that of Brown (40). For Homo sapiens, oxygen was breathed in through a tube from a reservoir of air whose oxygen supply was continually replenished, and breathed out through another tube. The exhaled breath passed through two large K O H bottles, then through a sampling flask before it was rebreathed. Two rubber balloons attached to the K O H reservoirs enabled the gas volume of the system to expand and contract with each breathing cycle and made it pos- sible to make quick estimates of the amount of additional oxygen required. After a large fraction of the oxygen had been consumed, the isotopic com- position of the unreacted oxygen was determined in a mass spectrometer. The fractionation factor a was then calculated from the equation MALCOLM DOLE, Natural History of Oxygen 21 --m/ a _ y - -yoe (31) O/ Xo ~ xo e - m / a where y is the a tom percentage of 0 is in the sample at the end of the experi- ment, y o is the 01 s percentage in the reservoir at the start of the experiment, Xo is the O is percengage in the oxygen added to the reservoir during the experiment, and rn is equal to the ratio of oxygen consumed to the amount initially present. If much oxygen is respired so that m / a is equal to 3 or 4, the exponential terms can be neglected and a calculated directly from the ratio y / x o . The results of this work are illustrated in Fig. 3 where the dotted I I I '0--¢ I I I [ © - ~ o I I ooo I I t i I I I I I I I I I I I I ~ 0 0 - - 00~ 0 [ ~ 0 ~ 0 ~ I I I LO0 1.02 a HUMAN BEING GREEN LEAVE~ CRAB FROG VEGETABLES FOREST LITTER MUSHROOMS MOLDS BACTERIA 1.04 1.06 FmtmE 3. Oxygen isotope fractiona- don factors observed during respiration or organic oxidation. vertical line represents the fractionation factor necessary to account for the enhanced O is content of the atmosphere. As shown by Dole, Hawkings, and Barker (41) the equation describing the changes per year in the oxygen isotope ratio of atmospheric oxygen is N 1 - - No = An [ r ~ - ri] [l + Nd ?/o,16 where N1 = [018]/[016] in atmosphere at end of year. No = [018]/[01~] in atmosphere at start of year. (32) 2 4 STRUCTURE AND FUNCTION OF OXYGEN the latter is an inadequate explanation of the enhanced O 18 content of the atmosphere. The theory of Lane and Dole (39) that fractionation of the oxygen isotopes by respiration and organic oxidation causes the increase in the O 18 content of the atmosphere seems to be the best from a quantitative standpoint. E. The Oxygen Isotope Cycle in Nature We can now set up the following steady state cycle of the oxygen isotopes in nature: Photosynthesis / / . ~ Greater O16 y i e l d - ~ . ~ A Land and tmosphere Respiration and organic oxidation Greater O 1~ consumption In summary, photosynthesis yields oxygen containing a higher O16/O 18 ratio than the oxygen of the atmosphere, while respiration consumes oxygen containing a higher O18/O I s ratio than the oxygen of the atmosphere and the same ratio as that of photosynthetic oxygen, thus leading to the non-equi- librium steady state value of the O16/O 18 ratio in the atmosphere. In other words, the O le ratio of the atmosphere has been automatically adjusted to just the right value so that the ratios for photosynthetic oxygen delivered to the atmosphere and the oxygen extracted from the atmosphere by respiration and organic oxidation are equal. VII . T H E F U T U R E O F A T M O S P H E R I C O X Y G E N Berkner and Marshall (2) suggest that during periods of high rates of photo- synthesis without much organic matter to consume the oxygen produced, the oxygen may have "overswung" its present level of abundance to a some- what higher value. During the ice ages of the Permian period, the earth cooled, photosynthesis was reduced "leading to a radical loss of atmospheric oxygen." Thus, it is highly likely that the oxygen content of the atmosphere has slowly fluctuated about a steady state concentration. Furthermore, in- asmuch as the H atom escape must still be continuing, even if only at a very slow rate, the net oxygen content of the atmosphere may be slowly rising. Unfortunately, the problem of determining experimentally the atmospheric oxygen percentage over millions of years in the past seems to be insuperable. MALCOLM DOLE, Natural History of Oxygen 2 5 The author acknowledges gratefully stimulating conversations with Professor Paul Harteck, Dr. Bertram Donn, and Dr. Walter Weller and the gift of pertinent reprints and manuscripts from these authors as well as from Dr. L. C. Marshall, Dr. D. L. Gilbert, Dr. B. J. Mair, Dr. F.~O. Rice, and Dr. F. C. Schmehl. REFERENCES I° ~-~LLOR, D. P,, Origin and evolution of chemical elements, Chemistry, 1964, 37, No. 4, 12. 2. BERKNER, L. V., and MARSHALL, L. C., The history of oxygenic concentration in the earth's atmosphere, Discussions Faraday Soc., 1964, 37,122. 3. AL~R, L. H., The Abundance of the Elements, New York, Interscience Pub- lishers, Inc., 1961, 240. 4. GIAUQtm, F. W., and JO~mTON, H. L., An isotope of oxygen, mass 18. Interpreta- tion of the atmospheric absorption bands, J. Am. Chem. Soc., 1929, 51, 1436; Isotope of oxygen of mass 17, in earth's atmosphere, J. Am. Chem. Soc., 1929, 51, 3528. 5. DOLE, M., The chemistry of the isotopes of oxygen, Chem. Rev., 1952, 51,263. 6. CAmmON, A. E., and WmHBRS, E., Report of the International Commission on Atomic Weights (1961), J. Am. Chem. Soc., 1962, 84, 4175. 7. KUmER, G. P., Planetary atmospheres and their origin, in The Atmospheres of the Earth and Planets, (G. P. Kuiper, editor), The University of Chicago Press, 2nd edition, 1952, 306. 8. ALLEN, C. W., Astrophysical Quantities, University of London, Athlone Press, 1963, 151. 9. CHAMBERLIN, R. T., Geological evidence on the evolution of the earth's atmos- phere, in The Atmospheres of the Earth and Planets, (G. P. Kuiper, editor), The University of Chicago Press, 2nd edition, 1952, 248. 10. RuB~.Y, W. W., Geological history of sea water, Bull. Geol. Soc. Am., 1951, 62, 1111. 11. MILLER, S. L., The formation of organic compounds on the primitive earth, Ann. New York Acad. Sc., 1957, 69,260. 12. ABELSON, P. H., Some aspects of paleobiochemistry, Ann. New York Acad. Sc., 1957, 69,274, 276. 13. BROWN, H., Rare gases and the formation of the earth's atmosphere, in The Atmospheres of the Earth and Planets, (G. P. Kuiper, editor), The University of Chicago Press, 2nd edition, 1952, 258. 14. DOLF., M., The history of oxygen, Science, 1949, 109, 77. 15. HARTECK, P., and J~.NS~.N, J. H. D., Uber den Sauerstoffgehalt der Atmosphtire, Z. Naturforsch., 1948, 3a, 591. 16. PAETZOLD, H. K., Vertical atmospheric ozone distribution, Adv. Chem. Series, 1959, 21, 209. 17. SW~NOS, P., and MEIm~L, A. B., The spectra of the night sky and the aurora, in The Atmospheres of the Earth and Planets, (G. P. Kuiper, editor), The Uni- versity of Chicago Press, 2nd edition, 1952, 159. 18. KENNARD, E. H., Kinetic Theory of Gases, New York, McGraw-Hill Book Co., 1938, 83. 26 STRUCTURE AND F U N C T I O N OF OXYGEN 19. HARTECK, P., and REEves, R. R., Chemical reactions in the atmosphere, Angew. Chem., Internat. Ed. Engl., 1962, 63, 25. 20. BOUgDEAU, R. E., and BAUER, S. J., Structure of the upper atmosphere deduced from charged particle measurements on rockets and the Explorer VIII satel- lite, in Space Research III, Proc. 3rd Internat. Space Sc. Syrup., Washington, April 30, 1962, Amsterdam, North-Holland Publishing Co., 1963, 173. 21. URn.Y, H. C., The atmospheres of the planets, in Astrophysics. III. The Solar System. Encyclopedia of Physics, (S. FRigge, editor), Berlin, Springer, 1959, 52,363. 22. HENDERSOn, L. J., The Fitness of the Environment, New York, The Macmillan Company, 1913. 23. Private communication from B. Donn. 24. Inn, E. C. Y., and TANAKA, Y., Ozone absorption coefficients in the visible and ultraviolet regions, Adv. Chem. Series, 1959, 21,263. 25. Um~v, H, C., BRICKW~DDE, F. G., and Mm~vHY, G. M., A hydrogen isotope of mass 2 and its concentration, Physic. Rev., 1932, 40, 1. 26. WASHEUan, E. W., and Uva~Y, H. C., Concentration of the H ~ isotope of hydrogen by the fractional electrolysis of water, Proc. Nat. Acad. Sc., 1932, 18,496. 27. LEwis, G. N., The isotope of hydrogen, J. Am. Chem. Sot., 1933, 55, 1297. 28. Um~Y, H. C., and GREIFV, L. J., Isotopic exchange equilibria, or. Am. Chem. Soe., 1935, 57,321. 29. DOLE, M., The relative atomic weight of oxygen in water and in air. 3". Am. Chem. Soc., 1935, 57, 2731. 30. DoT.~., M., The relative atomic weight of oxygen in water and in air. A dis- cussion of the atmospheric distribution of the oxygen isotopes and of the chemi- cal standard of atomic weights, or. Chemic. Physics, 1936, 4,268. 31. DOLE, M., The chemistry of the isotopes of oxygen, Chem. Rev., 1952, 51, 263. 32. DOLE, M., and SLOBOD, R. L., Isotopic composition of oxygen in carbonate rocks and iron oxide ores, J. Am. Chem. Sot., 1940, 62,471. 33. DoI~, M., LAne, G. A., R~-~D, D. P., and ZAUK~TrES, D. A., Isotopic composi- tion of atmospheric oxygen and nitrogen, Geochimica et Cosmochimica Acta, 1954, 6, 65. 34. KmSI-mNBAUM, I., Physical Properties and Analysis of Heavy Water, New York, McGraw-Hill Book Co., 1951, 418. 35. DOLE, M., and JV.NKS, G., Isotopic composition of photosynthetic oxygen, Science, 1944, 100,409. 36. VInOORADOV, A. P., KUT'~RIn, V. M., UTUBEKOVA, M. V., and ZADOROZHNY, I. K., Isotopic composition of photosynthetic oxygen, Dokl. Akad. Nauk SSSR, 1959, 125, 1141. 37. VINOGR~a~OV, A. P., KUTYURIn, V. M., and ZADOROZHNY, I. K., Fractionation of the isotopes of atmospheric oxygen, Dokl. Akad. Nauk SSSR, Geochemistry Section, No. 3, 1959, 195. 38. RAKESTRAW, N. M., RtmD, D. P., and DOT.E, M., Isotopic composition of oxygen in air dissolved in Pacific Ocean water as a function of depth, J. Am. Chem. Soc., 1951, 73, 2976.